Faculty Bibliography

Cell-mediated cytotoxicity involving natural killer cells requires contact between effector and target cells for effective cytolysis. Ultrastructural studies of biopsies of primary human malignant melanoma showed mononuclear leukocytes to be located in close proximity to tumor cells and, on occasion, within the confines of the melanoma cell itself. This phenomenon, called emperipolesis, was examined in vitro to determine whether the same population of cells that exhibits emperipolesis is responsible for cytotoxicity. Since natural killer cells have been identified morphologically and functionally as large granular lymphocytes with surface receptors for the Fc portion of immunoglobulin (FcR+ cells), lymphocytes were depleted of FcR+ cells, and their cytotoxicity and ability to emperipolese were measured. Both of these properties were markedly diminished (88 and 85 per cent, respectively). Systematic comparison of emperipolesis and cytotoxicity from donors known to exhibit either high or low lymphocyte cytotoxicity showed perfect concordance. Ultrastructural analysis of in vitro emperipolesis revealed the emperipolesing lymphocytes to be FcR+ cells establishing identity with the large granular cells known to mediate cytotoxicity. The morphologic marker found in both FcR+ cells, purified by rosetting techniques, and in emperipolesed lymphocytes consisted of cytoplasmic parallel tubular arrays. Further studies designed to elucidate a mechanism for cytotoxicity and emperipolesis implicated cell surface proteases as mediators of these activities. Competitive inhibition of surface proteases with artificial and natural inhibitors markedly reduced both cytotoxicity and emperipolesis. Therefore, it is likely that lymphocytes that are FcR+ participate in cell-medicated cytotoxicity through mechanisms involving cell contact and enzyme-initiated damage of target cells. Emperipolesis represents one type of effector-target cell contact leading to cytotoxicity

Since it is well established that cellular immunity plays a role in the defense against melanoma, the morphologic aspects of this reaction warranted investigation. Accordingly, peripheral blood mononuclear cells obtained from healthy donors were incubated with human melanoma cells for 1 to 24 hours to examine, on the ultrastructural level, the cellular interaction that eventuates in cytolysis of the tumor cells. Within 1 hour of incubation, monocytes and lymphocytes were seen attached to approximately 40 per cent of the melanoma cells with marked interdigitation of cellular processes. After 4 hours of incubation, the percentage of tumor cells with attached leukocytes remained the same, but 2 to 9 per cent of the melanoma cells showed interiorized lymphocytes when kept in suspension, 10 to 25 per cent when maintained in culture dishes. Erythrocytes or fixed lymphocytes were not taken up by the melanoma cells nor were living lymphocytes seen in fibroblasts or endothelial cells which served as controls for the neoplastic cell lines. Thus, melanoma cells did not prove to be randomly phagocytic, and the interiorization displayed by lymphocytes--a process called emperipolesis--appears to be selective. It is postulated that emperipolesis may enhance the tumoricidal effect exerted by cytotoxic lymphocytes on melanoma cells

Human blood lymphocytes activated in vitro with antigen to which the donor is reactive are capable of suppressing the secondary proliferative response of autochthonous fresh cells to antigen. Both antigen-specific and antigen-nonspecific suppression can be detected in each experiment. These suppressor cells act by decreasing the number of lymphocytes entering the proliferative response rather than by slowing or otherwise inhibiting ongoing proliferation. The suppressor cells must be added soon after fresh cells are stimulated with antigen to be effective, but the suppressor cells themselves need not proliferate to exert their effect. Suppressor cells are optimally effective when added in numbers equal to those of the responding population, but still exert a significant effect at one-eighth that number

After our initial report tha leukocyte dialysates containing transfer factor augment the thymidine incorporation of antigen-stimulated lymphocytes, we have adapted the system to microleukocyte cultures. This modification permits both (a) the simultaneous assay of a single dialysate on the cells of multiple individuals, and (b) the assay of multiple dialysates on the cells of a single individual. The data thus secured, demonstrate that dialysates from both skin-test-positive and -negative donors produced similar degrees of augmentation whether the data are expressed as an arithmetic difference or as a ratio. When expressed as an arithmetic difference, the amount of augmentation is increased in proportion to the level of thymidine incorporation of the assay cells when they were stimulated by antigen alone. When expressed as a ratio, however, the degree of augmentation is independent of the response of the assay cells. An analysis of the ability of dialysates to engage previously uncommitted lymphocytes and thus to augment thymidine incorporation, revealed that precommitted cells were required. In these experiments, antigen-reactive cells were deleted from populations of peripheral blood lymphocytes by incubation with purified protein derivative of tuberculin, diphtheria toxoid, or streptokinase-streptodornase in the presence of [3H]thymidine of high specific activity. This deletion depressed or abolished the effect of dialysate on the residual population when it was recultured with the same antigen, but the effect on the response of the remaining lymphocytes to other antigens was unaltered. In this study, leukocyte dialysate appeared to augment nonspecifically the thymidine incorporation of an antigen-specific precommitted clone of lymphocytes. The relationship of these adjuvant effects on peripheral blood lymphocytes in vitro to the specific and nonspecific activities of transfer factor in vivo remains to be elucidated